Tobacco use is substantially higher among sexual minorities than among heterosexuals. The reasons for this persistent disparity remain unclear, but the high toll of death and disability from tobacco use creates substantial health inequalities in cancer. Same-sex couples' movement within the U.S. results in same-sex couples concentrating in more urban neighborhoods where there may be more tobacco retailers. Living near a tobacco retailer has been associated with decreased success in quitting tobacco. The presence or absence of overlapping patterns of same-sex couples and tobacco retailers have not been explored. This study uses a unique and high quality list of 92,384 tobacco retailers from 97 U.S. counties with 18,034 census tracts and data from the U.S. Census to examine the relationship between same-sex couple density and tobacco retailer density in census tracts. The analysis uses geographical information systems (GIS) software that allows for buffering census tract areas to create a measure of tract tobacco retailer density. This relationship is tested usin spatial regression models to account for non-independence due to geographic proximity. The relationship is then examined to see if it is the result of other, confounding census tract characteristics. A multiple mediation approach is proposed to examine if the relationship between tobacco retailer density and same-sex couple density is confounded by tract characteristics such as urbanicity, income, and racial/ethnic composition.